01:05:54  * bradleymeckjoined
02:30:07  * bradleymeckquit (Quit: bradleymeck)
03:47:41  * Tweth-U-PDSjoined
03:50:22  * Tweth-V-PDSquit (Ping timeout: 265 seconds)
05:19:06  * decoderquit (Ping timeout: 244 seconds)
05:19:17  * decoderjoined
05:23:45  * plutonuxquit (Quit: Leaving)
07:06:32  * sof_quit (Quit: Connection closed for inactivity)
09:42:36  * mikolalysenkoquit (Ping timeout: 244 seconds)
09:42:59  * mikolalysenkojoined
14:08:06  * bradleymeckjoined
14:21:54  * seventhjoined
14:52:04  * plutoniixjoined
15:11:17  * arvquit
15:11:38  * arvjoined
15:13:58  * Guest59quit (Ping timeout: 268 seconds)
15:18:17  * cstipkovicquit
15:19:15  * cstipkovicjoined
15:42:45  * dagobert__quit
15:43:08  * dagobert__joined
16:24:59  * RT|Chatzillaquit (Read error: Connection reset by peer)
16:27:00  * bradleymeckquit (Quit: bradleymeck)
17:22:04  <trungl-bot`>Tree closed by [email protected]: Tree is closed (Automatic: "Check" on http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20nosnap%20-%20debug/builds/10689 "V8 Linux - nosnap - debug" from 292b3548f6d02b964b4afe3e05f89c0681fa5620: [email protected])
17:35:10  <trungl-bot`>Tree opened by [email protected]: Tree is open (Automatic: ¡Boa fortuna!)
17:43:50  * bradleymeckjoined
18:02:34  * evanlucasjoined
18:03:07  * evanlucaspart
18:17:52  * seventhquit (Ping timeout: 260 seconds)
18:37:45  * Guest59joined
18:43:15  * Guest59quit (Quit: Textual IRC Client: www.textualapp.com)
18:47:14  * Guest59joined
18:48:31  * Guest59_joined
18:52:53  * Guest59quit (Ping timeout: 268 seconds)
19:02:23  * plutoniixquit (Read error: Connection reset by peer)
19:02:58  * plutoniixjoined
19:10:33  * seventhjoined
19:11:30  * seventhquit (Client Quit)
19:34:29  * kenansul-quit (Quit: ZNC - http://znc.in)
19:35:53  * kenansulaymanjoined
19:54:12  * sxaquit (Ping timeout: 256 seconds)
19:55:17  * sxajoined
20:32:47  * decoderquit (Ping timeout: 268 seconds)
20:37:02  * decoderjoined
21:04:58  * xaxxonjoined
22:30:38  * bradleymeckquit (Quit: bradleymeck)
22:33:23  <caitp>aklein: so yeah, the big win seems to come from using named loads rather than keyed loads. doesn't matter for turbofan since it will optimize monomorphic cases the same way, but in the baseline case, keyed loads look like more work
22:33:44  <caitp>we could probably change it so that the AST desugaring gets that benefit, but userspace js won't ever see a benefit from it
22:34:09  <caitp>either way, desugaring in the bytecode generator is a lot nicer and easier to understand imo
22:42:28  * RT|Chatzillajoined
23:03:47  * Keverwquit (Quit: Textual IRC Client: www.textualapp.com)
23:03:55  <aklein>caitp: yeah, thanks for the investigation, I agree that since we have to change the AST anyway, doing the desugaring in the bytecode generator seems reasonable
23:05:19  * Guest59_quit (Quit: My Mac has gone to sleep. ZZZzzz…)
23:06:44  <caitp>once we are desugaring destructuring via BytecodeGenerator instead of the ast, adding new "special" ast nodes like that will probably be less scary
23:10:02  * Guest59joined
23:25:27  <aklein>agreed
23:25:31  <aklein>will respond on the CL
23:28:00  * Guest59_joined
23:31:38  * Guest59quit (Ping timeout: 250 seconds)